Earth and the Internet

Although the main plot-line concerns a black hole gone amok and threatening the physical integrity of the planet, Earth is far more than the usual "scientists-trying-to-save-the-world-from-things-that-Man-should-not-meddle-with" stuff of B-grade science fiction (both books and movies). Brin makes interesting ecological and cultural extrapolations from the present trends in the world as they may be modified by the history of the next four decades. He explores issues of vital interest to librarians, such as privacy, the future of publishing, access to information, and copyright.

In Earth, the net is the primary means of communication -- mail, education, publication, entertainment, and most forms of social interaction. Traditional publishing has completely ceased, to the point where one of the characters, confronted with the need to read a book for the first time in his life, finds the experience difficult:

"If only it were a modern document, with a smart index and hyper links stretching all through the world data net. It was terribly frustrating having to flip back and forth between the pages and crude, flat illustrations that never even moved! Nor were there animated arrows or zoom-ins. It completely lacked a tap for sound.

"... in a normal text you'd only have to touch an unfamiliar word and the definition would pop up just below. Not here though. The paper simply lay there, inert and uncooperative."[1]

The Internet is a system of interconnected computer networks, which share an addressing system and communications protocols (ways of doing things) so that a person using one computer network can easily communicate with people using another computer network. Ordinarily, each computer network has its own commands and ways of doing things. For example, the command to get rid of a message permanently might be "purge" on one computer or network of computers wired together, while it might be "erase" on another. Human beings understand the concept of synonyms easily, but computers are more limited. Rather than try to program in Roget's, computer programs and computer networks have been set up to respond in predictable ways to specified commands. In the past, the problem has been that the commands were different for each piece of software.

Similarly, electronic (e-mail) mail systems are generally set up to store messages for one person that are entered by another person using the same computer or network of computers. They are analogous to a small post office where the patrons must bring letters the office and the postal employees place them in a pigeon-hole for another patron to pick up. This system, with no connection to the "outside" world, works well as long as all the patrons are willing to come to the central post office to pick up their messages and as long as they don't want to communicate with anyone who doesn't have a local address. As long as the employees understand the filing system, it doesn't fundamentally matter whether the pigeon- holes are arranged alphabetically, by street address, or by Social Security Number.

To extend the analogy, when the postal patrons start wanting to communicate with people who use other post offices, both a system of distribution for the mail to get from one office to another, and a uniform system of addressing are needed. Without the distribution system, the mail will sit where it is first deposited. Without uniform addressing, the mail may possibly arrive at the destination post office, but it will be unclear to whom it is intended that it be delivered. If each town has arbitrary codes to designate the other towns, it will probably not even make it to the post office level.

The TCP/IP (Transmission Control Protocol/Internet Protocol) suite is a group of computer protocols designed to allow many different types and brands of computers and computer networks to communicate with each other. Originally designed to facilitate communication among contractors for the U.S. Department of Defense, the protocols were used in the 1970s and early 1980s to run the Advanced Research Projects Agency network (ARPANET). ARPANET was originally designed to allow the contractors to share expensive computer resources that many of them needed for a limited time on their projects (such as databases and the then very expensive graphics programs), but it quickly became a communication network as the people working on such projects found it a convenient way to communicate and collaborate electronically.

As time went on, ARPANET became too popular for the capacity of the hardware upon which it was based and too big for the addressing system provided for in the protocols. Addresses require a specified amount of computer memory; the network grew faster than the original addressing scheme had ever anticipated. It is similar to using the letters of the alphabet as addresses -- if one-letter addresses are used, there can be twenty- six users before one runs out of unique identifiers. If two letters are used, then 676 unique addresses are possible, and so on. The addressing system has been expanded several times in the intervening years, and has grown into a layered system of (usually 3 letter) codes designating the network, sub-network, and particular local machine.

By itself, ARPANET was a large network. However, as more and more vendors and others working on government projects joined it, ARPANET became more than just a large network (a group of computers connected to each other) and took on more and more of the characteristics of an internetwork (a group of interconnected networks). An internet is any network of two or more computer networks. The Internet is "an internetwork of many networks all running the TCP/IP protocol suite, connected through gateways, and sharing common name and address spaces."[2] In other words, the Internet is a particular internet, based largely in the United States but with connections world-wide, which allows computers at government agencies, educational institutions and corporations to communicate with each other and which facilitates resource-sharing and communication among the people who are able to log on to any of its constituent computers (called "hosts" or "nodes").

Another network, used heavily by educational institutions, is BITNET. There are gateways (computers that provide linkages) between BITNET and the Internet, but BITNET is not, properly speaking, part of the Internet. BITNET has its own addressing scheme and channels its messages through dedicated phone lines physically connecting adjacent computers. It, like the Internet, provides for resource sharing and communication and collaboration among its member schools, but it is primarily limited to educational institutions.

Today, the Internet and BITNET provide easy, fast, and generally reliable communication to researchers and scholars all over the world. Through the Internet, the library catalogs of an ever-growing number of college and university libraries are accessible. If an institution is going to have an online catalog and is going to allow people to dial in and access it by modem, it is a relatively simple thing to allow access over the Internet; many institutions have done this. Through BITNET and the Internet, discussion groups can be hosted on mainframes which allow, for example, catalogers in Richmond, VA to communicate with catalogers in Hong Kong, easily, economically, and almost instantaneously (certainly much faster than normal mail). Since the connections between the computers in the various networks and to the Internet are subsidized by the owners of the mainframe computers that are networked together, the cost to the individual person using the Internet is usually at most the price of a local telephone call.

To make the leap from the Internet as it exists today to the situation that David Brin conjures up in Earth could easily be accomplished in forty years. Some of the problems that exist on the Internet now in minor form have grown to problems of major proportions in Earth but at a rate that is believable in forty years. For example:

In Brin's vision, privacy even as a polite fiction is completely dead. One of the characters essentially commits suicide because what he had considered to be a private conversation was recorded and included in memoirs published on the world data net, and he has no other legal or moral recourse. Crime is virtually non-existent, because anyone with $8.50 to spend can buy a pair of "true-vue sensu-record goggles, with net access."[3] In an overpopulated society, almost any human interaction has been recorded by some bystander, who is only too happy to fax it to the police.

Although the world data net e-mail and private storage areas are supposed to be private, anyone with the necessary programming skills can steal access to another's private data. In fact, there is a cottage industry engaged in doing just that and another one engaged in defending data. However, there are apparently no laws to punish offenders in this area, and the societal taboo against secrecy regards expressed desires for privacy with suspicion.

Given the premise of the Helvetian War and the hatred of secrecy it engendered, this lack of concern for privacy is believable in the story. Given that wide use of computer networks is in its infancy and that the problem will only become worse as more people gain access to the Internet, now is the time to be working on solutions to the problem, or at least etiquette to handle faux pas gracefully.

In Earth, publication is a relatively simple process. One creates a work, usually multi-media, and makes it available on the net. There is a system for automatically notifying anyone living who is mentioned in the work, but there seems to be no requirement that their permission be given before recorded images of them are included in the work.

Computer memory space is also not a problem. It is cheap and full-text (or image) storage is no problem. There are charges for access to some sources of information (such as real-time satellite images of disaster areas), but there seem to be no restrictions on who can access what and no practical limit on either an individual's or on the system's storage capacity.

Indeed, the problem isn't gaining access to information. The problem is screening out the irrelevant bits of information. There are artificial intelligence "personalities" who act as secretaries to screen the electronic mail (and answer the routine stuff) and there are programs called "ferrets", "foxes", and "bloodhounds" that people can send out into the net to retrieve information on whatever it is they want to know about. [Ed.: Have you used your personal information manager or checked with your gopher today?] Although these programs are available "off the shelf", the more effective ones are custom designed. Relevant information is one of the "perks" of the computer literate -- a level of computer literacy that is largely limited to professional programmers in our world.

With all of this information, there appears to be little or no concern about the "pre-net" publications. One of the characters runs into an essay that is only available in "antique" hard copy form but, with such a mass of readily available information, it appears that no one worries about the unconverted works that are gradually being lost. Similarly, it seems to bother no one that new editions of a work can appear almost daily, and there is no evidence in the novel that anyone is particularly concerned about preserving the old editions.

We are facing some of this problem today, with the proliferation of electronic journals. We are grappling with the problem of how to provide access to them, who should preserve them in their various editions, and whether and for how long archives of discussion lists should be kept. Computer professionals are primarily concerned with backing up data so that it can be restored on another computer if the system goes down. They are not, by and large, concerned with preserving the data for scholars to track all the different forms that it appeared in. In Earth, when the system is temporarily disrupted, people react as we would if the phone service was temporarily interrupted. They do not appear to be concerned that the sum total of human knowledge (from their perspective) was in danger.

The book raises the issue of ease of plagiarism and copyright violation as well. As anyone with a word processor or a graphics editor knows, text can be changed, added to, deleted; graphic images stored in a computer can be changed ("enhanced") to do things like altering photographs undetectably. At the present time, the Gutenberg Project (a project to provide machine- readable text for literary "classics") has chosen to deal with this issue by only converting works that are out of copyright. There is nothing now (except ethics enforced by public opinion) to keep someone from "enhancing" Alice in Wonderland and presenting it as their own work. [Ed.: My electronic version of Alice is copyrighted by the person who wrote the introduction and added some notes.] What will happen when the work of living people starts being liable to "enhancement"? Particularly undetectable "enhancements"?

A related question, more directly tied to the growth of the Internet, is: what will happen as the desire of data vendors to restrict usage of the data (so that they can sell it to more customers) collides with the desire of libraries to provide access to information over the Internet?

There is nothing about the net in Earth that is inherently improbable. Brin provides us with a view of a possible future of the Internet and the networks which will succeed it. It is a wonderful piece of speculative fiction that raises interesting questions for librarians, "computer people", and society at large.

1. David Brin, Earth. (New York: Bantam, 1991). p.199

2. John S. Quartermain, The Matrix. (Bedford, MA: Digital Press, 1990).

3. Brin, p.308.

4. Brin, p.468.